Superposition of local zoom models and simultaneous calibration for 1D-2D shallow water flows

We address the problem of coupling 2D shallow water equations with 1D shallow water equations (St-Venant equations), as applied to river-floodplain flows. Mathematical coupling conditions are derived classicaly from the 3D Navier-Stokes equations by integrating over the vertical wet section, when overflowing occurs. It leads to extra source terms in the I D equations. Next we assume to be in a variational data assimilation context, then the optimal control process allows to couple both models and assimilate data simultaneously (Joint Assimilation Coupling algorithms). Two different versions of JAC algorithms are presented and compared. In a numerical test case, we superimpose the local 2D model on the I D global model. The results show the efficiency of the present simultaneous superposition-assimilation approach. (C) 2009 IMACS. Published by Elsevier B.V. All rights reserved.